113 ms·
I hope everyone recognized the cheap tokens as a transparent ploy to gain more users before jacking up prices. This was obvious from day one.
by solenoid0937 9d ago
I hope everyone recognized the cheap tokens as a transparent ploy to gain more users before jacking up prices. This was obvious from day one.
- brazukadev 9d agoAs long as there is no path to profitability going from subsidized to fully paid tokens we are making them lose money.
- tehjoker 9d agoThis is not entirely true, you are costing them in the short run, but are growing your dependence on their product and feeding them training data. Their goal is to create AGI and replace humans with something resembling the plot of Horizon Zero Dawn and its sequel but more extreme.
- brazukadev 4d agoThat's their goal but if they fail we got LLMs "for free".
- password54321 9d agoIt is really about data. The endgame is not about hoping devs give them money.
- copperx 9d agoMore data isn't helping much, if Astra is any indication.
- password54321 9d agoThis is just you putting your head in the sand while you gradually automate your job.
- flir 9d agoI've spent my entire career trying to automate my job. (I get what you're trying to say, I'm just finding the irony amusing).
- mrweasel 9d agoWhat are they going to do with that data? Is it ads again?
- password54321 9d agoYou are about to get replaced.
- solenoid0937 9d agoThat's happening anyways, they don't need more organic data for that. Also, who cares? Such is the price of progress, I'm not selfish enough to hold humanity back because "muh job"
- password54321 9d agoThat's great. We are obviously both privileged enough to think that. Not so great for others.
- solenoid0937 9d agoThey'll get through it. You don't just halt progress because the transition will be rough. We'd be pre-industrial if we thought like this.
- mrweasel 9d agoProbably not, but I can channel my inner Scotty Fairno https://www.instagram.com/reel/DUbMs8Vipp5/ https://www.instagram.com/reel/DUbMs8Vipp5/
- geraldwhen 9d agoThey won’t be able to. Open models exist and run on a Mac you can buy now or in October.
- apparent 9d agoThe audience for this sort of thing is currently quite limited. It may grow more in the future, especially if prices get out of control, but for now, it's a relatively small subset of people.
- howdareme9 9d agoplease point me to an open model, that matches astra even on low reasoning, and runs on a mac
- taylorfinley 9d agoNo problem, but you'll need to wait a few months
- dwaite 9d agoWhy would it need to match Astra? Software companies don't give up and fold because they can't hire one of the five best engineers in the world.
- 7bit 7d agoWhat kind of nonsense argument is this
- esskay 9d agoSorry but this is a pretty delusional take. No, you cant run the same level or even close to that level of model on a mac, not even on a 512gb studio. You can run good models sure, but not models anywhere close to the capabilities of these ones yet, despite what idiots on Twitter keep spouting.
- andybak 9d agoThe "run locally" thing is a red herring. The fact that anyone can be an inference providers and offer them as a service is what we need to focus on. If they are "good enough" then market forces will stop Anthropic/OpenAI from charging monopoly rents.
- fny 9d agoI hope everyone recognizes they aim to make you dependent on their intelligence rather than your own.
- catchnear4321 9d agoI reject your intelligence and substitute my own?
- zulux 9d agoThere's an escape valve: I've had Claude stand-up AI-enabled features in my app, so we're much less dependent on Claude itself. The app uses cheap API calls to check code quality and run other "lessons learned" sweeps. Some of these become regular code in the end as well.
- vidarh 9d agoSame. Claude "loves" writing tools to make checks deterministic so much that I actually have to pull it back sometimes and point out that some things need to keep using LLMs. But even then, I have it "farm out" a lot of things to scripts using dirt-cheap models for things I don't need a Sonnet or higher level model for as well. The proportion that runs on the SOTA models keeps dropping.
- gruez 9d agoBut there's open models that are (pessimistically) 1 year behind? It's not like if openai decided to rugpull everyone, we'd be going back to writing code by hand and all the developers who forgot how to fizzbuzz would be screwed.
- benterix 9d ago> we'd be going back to writing code by hand Well, as as SWE who writes code by hand and has no intention to outsource their intelligence to any entity, I felt slightly offended by "back".
- 9d ago
- flir 9d agoOr more efficient models/faster hardware close the gap between here and profitability. I wouldn't be surprised if every $10 you add to the subscription cost halves your audience.
- mrweasel 9d agoThat boils down to who the customers for ChatGPT really are. $10 isn't going to make a difference for those who already pay, because they are basically pro-consumers. OpenAI might deliver enterprise solutions for large companies, but those deals are besides the per month subscriptions on the website. For ChatGPT I really only see two customer groups, pro-consumer and small and medium enterprise. The entry level are just going to use Google/Gemini for the most part, or some ad supported ChatGPT tier. There's no price low enough to make the entry level, average consumer pay for an AI service. These are people who will not pay for search, email and social media. They only pay for streaming because there's no way around it.
- flir 9d agoOk half was an exaggeration, but I think $10 might make quite a lot of them jump to Claude. Or DeepSeek. Remember, we're talking about a global user base here. I settled on GPT Plus/Codex for my own work, while my employer settled on Claude on Bedrock (where I am, according to their metrics, using about $20 of compute a day, so I can believe that GPT Plus is a loss-leader). If I hit too many 5-hour timeouts I'm not jumping to Pro I'm jumping to Claude. (Funnily enough, I just hit one. It'll reset in... three hours. I wasn't even hammering it, just writing some tests. I think I'm gonna just go reactivate my Anthropic account for a month).
- mrweasel 8d agoI'd agree that $10 can and will make users switch to a different provider. Just as streaming music services are all price basically the same, as to avoid users just jumping to the cheapest, I think the AI companies will find a suitable price point for consumer AI. After that they'll just settle into a system where when one raise or lower their price, the rest will follow. Those who benefit from using and LLM will continue to pay for one, regardless of provider, regardless of the price being $20 or $50. The global aspect is interesting, because $50 is a lot of money for most of the global population. The AI advocates claim to be getting more hundreds, if not thousands, of dollars of value from their subscription. In the poorest countries that should make an even greater difference, but that assumes that they can get those first $50 and that the value is actually being generated.
- skybrian 9d agoYour cheap cynicism makes OpenAI's lower prices sound like a reason to avoid using them, but that's backwards! It's a better strategy to use them now while prices are low and switch later if necessary. Nobody knows who will have the best prices later, so better to decide later.
- cj 9d agoIt’s good to be aware when you’re buying something that’s subsidized. No need to downplay the importance of awareness.
- skybrian 9d agoAware of what, though? We don't actually know OpenAI's profit margins on inference. Do we need to know? It's enough to know that the market is competitive and new models with better prices are released often. This means you should have a way to switch models.
- cj 9d agoThe entire market is subsided more than it is competitive. The market wouldn’t exist the same as it does today if it weren’t being propped up. Thats the point. I remember the days when I had a $2000+ credit balance on Uber because I mastered their referral program. “Free” uber blacks for 2 years! When you’re using free shit that you should probably be paying more for, it’s good to not fool yourself. In the Uber example, the fact that Lyft also existed at the time didn’t matter to me as someone happily riding around the city in Uber Black for $0. Unfortunately Lyft’s referral program wasn’t so lucrative so I couldn’t switch…
- fgkramer 9d agoHow does this work with inference providers behind OpenRouter? Are you saying that all of them are actually losing money?
- 9d ago
- madaxe_again 9d agoOr to just give users enough usage to find categories in which it is useful for them. I pay the $200/mo., and don’t regret it for an instant - I have a project manager, an executive assistant, a business analyst, a software developer and an international accountant, for an absolute song.
- dgellow 9d agoyou’re using an LLM as your international accountant? And brag about it online? That’s definitely a choice
- TuxSH 9d agoIt was obvious indeed, but damn does it feel good to have ridden the May 25 - Aug. 26 gravy train. Now, let's see how the Anthropic IPO goes.
- edoceo 9d agoI'm going to predict: soft, below first-trade price at 180d from IPO date.